prerequisites

Prerequisites

To follow the coding examples in the workshop, you can either use a Visual Studio Code Dev Container, which will run an embedded Kubernetes cluster, or you can install the softare locally.

The minimum hardware requirements for the tutorial are:

• Quad-core CPU
• 8 GB of RAM

Option A: VS Code Dev Container

Please install the following software:

• Docker
• Visual Studio Code
• VS Code Dev Containers Extension

1. Start the Dev Container

1. After cloning this repository, open it is Visual Studio code.

2. A notification should appear, asking if you want to "Reopen in Container". Click on that button to reopen the workspace in a dev container. If the notification does not appear, press F1 and search for "Dev Containers: Reopen in Container".

3. Building the container image and starting the container takes several minutes. It involves pulling various container images, starting a Minikube Kubernetes cluster and installing Prometheus in it. You can open the log to see the current progress.

4. Once the setup has completed, you can start working. To stop the container, simply close VS Code and the container will stop a few seconds later.

2. Deploy the Test Workload

This workshop uses the Kubernetes resource-consumer as the target workload. To deploy it, open a terminal in the dev container run the following command:

kubectl apply -f ./prerequisites/resource-consumer.yaml

Option B: Local Setup

Please install the following software:

• Docker
• Node.JS v18 or higher
• minikube
• kubectl
• helm

1. Set up a Local Kubernetes Cluster

1. Configure minikube to use 4 CPUs and 4 GB of RAM for the cluster node. Minikube offers various methods fro creating a node, called drivers. The following snippet assumes that you are using Docker as the driver, but you may change to another one, if desired.

   minikube config set cpus 4
   minikube config set memory 4096
   minikube config set driver docker

2. Create a cluster:

   minikube start --kubernetes-version=v1.27.4

3. Verify that the cluster is running:

   kubectl get nodes

2. Install Prometheus

1. Add the Prometheus helm chart repository to your helm configuration:

   helm repo add prometheus-community https://prometheus-community.github.io/helm-charts
   helm repo update

2. Install the kube-prometheus stack in the monitoring namespace. This may take a few minutes.

   kubectl create namespace monitoring
   helm install prometheus prometheus-community/kube-prometheus-stack \
     --set namespaceOverride=monitoring --set grafana.namespaceOverride=monitoring \
     --set kube-state-metrics.namespaceOverride=monitoring \
     --set prometheus-node-exporter.namespaceOverride=monitoring

3. Wait for all Prometheus pods to be ready. You can check this with the following command:

   kubectl get pods -n monitoring

4. Check if the Prometheus UI is available by forwarding a local port to its service.

   # Forward local port 9090 to the Prometheus service
   kubectl port-forward -n monitoring services/prometheus-kube-prometheus-prometheus 9090:9090
   
   # Go to http://localhost:9090 in your browser.
   # If you see the Prometheus query UI, everything is fine.
   # Stop the port forwarding using Ctrl-C

3. Deploy the Test Workload

This workshop uses the Kubernetes resource-consumer as the target workload. To deploy it, run the following command in a terminal in this folder:

kubectl apply -f ./resource-consumer.yaml

4. Stopping and Deleting the Cluster

To temporarily stop the minikube cluster, run the following:

minikube stop

You can resume the cluster operation using:

minikube start

# Wait for all pods to be running again
watch kubectl get pods -A

To delete the cluster, run:

minikube stop
minikube delete